This research assessed the comparative risk associated with exposure to traffic pollution when travelling via different transport modes in Christchurch, New Zealand. Concentrations of PM1, UFPs and CO were monitored on pre-defined routes during the morning and evening commute on people travelling concurrently by car, bus and bicycle. It was found that car drivers were consistently exposed to the highest levels of CO; on-road cyclists were exposed to higher levels of all pollutants than off-road cyclists; car and bus occupants were exposed to higher average levels of UFP than cyclists, and travellers were occasionally exposed to very high levels of pollution for short periods of time. PM10 and PM2.5 were found to be poor indicators of exposure to traffic pollution. Studying Christchurch adds to our understanding as it was a lower density city with limited traffic congestion compared most other cities previously studied.

Recent progress in developing artificial neural network (ANN) metamodels has paved the way for reliable use of these models in the prediction of air pollutant concentrations in urban atmosphere. However, improvement of prediction performance, proper selection of input parameters and model architecture, and quantification of model uncertainties remain key challenges to their practical use. This study has three main objectives: to select an ensemble of input parameters for ANN metamodels consisting of meteorological variables that are predictable by conventional weather forecast models and variables that properly describe the complex nature of pollutant source conditions in a major city, to optimize the ANN models to achieve the most accurate hourly prediction for a case study (city of Tehran), and to examine a methodology to analyze uncertainties based on ANN and Monte Carlo simulations (MCS). In the current study, the ANNs were constructed to predict criteria pollutants of nitrogen oxides (NOx), nitrogen dioxide (NO2), nitrogen monoxide (NO), ozone (O3), carbon monoxide (CO), and particulate matter with aerodynamic diameter of less than 10 μm (PM10) in Tehran based on the data collected at a monitoring station in the densely populated central area of the city. The best combination of input variables was comprehensively investigated taking into account the predictability of meteorological input variables and the study of model performance, correlation coefficients, and spectral analysis. Among numerous meteorological variables, wind speed, air temperature, relative humidity and wind direction were chosen as input variables for the ANN models. The complex nature of pollutant source conditions was reflected through the use of hour of the day and month of the year as input variables and the development of different models for each day of the week. After that, ANN models were constructed and validated, and a methodology of computing prediction intervals (PI) and probability of exceeding air quality thresholds was developed by combining ANNs and MCSs based on Latin Hypercube Sampling (LHS). The results showed that proper ANN models can be used as reliable metamodels for the prediction of hourly air pollutants in urban environments. High correlations were obtained with R (2) of more than 0.82 between modeled and observed hourly pollutant levels for CO, NOx, NO2, NO, and PM10. However, predicted O3 levels were less accurate. The combined use of ANNs and MCSs seems very promising in analyzing air pollution prediction uncertainties. Replacing deterministic predictions with probabilistic PIs can enhance the reliability of ANN models and provide a means of quantifying prediction uncertainties.

This paper describes an in-depth analysis to investigate the huge variation in the measured roadside air-pollutant concentrations of carbon monoxide and nitrogen dioxide in terms of the traffic flow levels, the orientation of the street to the prevailing wind, the wind speed, temperature and barometric pressure. The work has attempted to develop generic parameters that can be applied to other urban areas. However, in the absence of a measure of congestion at the site in Palermo (Italy), the methodological approach proposed used the simultaneous noise measurements, in units of decibels (B), to help parameterise a generic congestion indicator in terms of the traffic flow. The potential transferability of the approach was demonstrated for a site in Marylebone Road, London (UK), given the similarity of the two study sites, canyon shape, traffic characteristics and road orientation. The results showed that, within the range of data available, noise levels could be used as a proxy for flow change on the shoulders of the peak hour and hence congestion and a generic relationship with factors statistically significant at 99 % confidence allows roadside concentrations due to traffic to be estimated with a regression coefficient of R (2) = 0.73 (R = 0.85). The research demonstrates that whilst there are indeed underlying relationships that can explain the roadside concentrations based on traffic and meteorological conditions, evidence is presented that confirms the complexity of the physical and chemical processes that govern roadside concentrations.